The utility of low-density genotyping for imputation in the Thoroughbred horse

Table 3 The mean correlation between true and predicted genotypes, as calculated in the within-population analysis of the UK dataset

Number of SNPs¹	Method for selection of low density SNPs
per individual	bpEQ	bpMAF	lduMAF
384	0.46 (0.14,0.89)	0.49 (0.20,0.91)	0.53 (0.22,0.89)
768	0.64 (0.36,0.93)	0.66 (0.38,0.93)	0.69 (0.37,0.94)
1K	0.70 (0.41,0.93)	0.78 (0.51,0.96)	0.75 (0.47,0.98)
2K	0.86 (0.53,0.99)	0.88 (0.62,0.98)	0.85 (0.52,0.99)
3K	0.92 (0.59,0.99)	0.94 (0.60,1.00)	0.88 (0.48,0.99)
6K	0.97 (0.73,1.00)	0.97 (0.71,1.00)	0.93 (0.61,1.00)
per SNP
384	0.30 (-0.17,1.00)	0.32 (-0.14,1.00)	0.36 (-0.08,1.00)
768	0.52 (-0.08,1.00)	0.53 (-0.06,1.00)	0.55 (-0.05,1.00)
1K	0.60 (-0.04,1.00)	0.67 (-0.05,1.00)	0.64 (-0.05,1.00)
2K	0.81 (-0.04,1.00)	0.83 (-0.02,1.00)	0.79 (-0.02,1.00)
3K	0.89 (-0.01,1.00)	0.90 (-0.03,1.00)	0.83 (-0.02,1.00)
6K	0.95 (0.25,1.00)	0.96 (0.49,1.00)	0.90 (-0.01,1.00)

Mean correlation between true and predicted genotypes per individual or per SNP for ECA1, with minimum and maximum values in brackets (tables for all chromosomes are in Additional file 7: Table S4); ¹total number of SNPs that would be on a genome-wide LDP of equivalent density.

ISSN: 1297-9686